TBA

Heraclitus famously observed that you cannot step into the same river twice, because it is ever-changing. In the same sense, historical Internet measurements are precious and irreplaceable. When future historians try to understand our times, will they still have the primary data sources to understand how the Internet grew, and how it became so central to the operation of global society?

In this talk, we will present an overview of the Internet's measurement history from an archival perspective, showing how Internet operators are uniquely positioned to preserve technical datasets that are in danger of being lost. We will introduce a new initiative, the Internet Data Trust, that seeks to identify and preserve Internet measurement as collective digital heritage. And we will identify some of the specific gaps that need to be filled in order to preserve our understanding of the history of the Internet in the Asia-Pacific region in particular.

This talk explores the practical realities of deploying IPv6 and transitioning away from IPv4 in enterprise networks. While we will cover the essential underlying concepts, the primary focus is on the "why, how, and what to expect" based on real-world success stories. We will examine the current state of transition technologies and how modern operating systems perform in the absence of IPv4. The session aims to provide a broader perspective on the implementation landscape, supported by actionable deployment guidance and resources for further reading.

The BGP routing environment provides a unique insight into the behaviour of the public Internet as a whole. BGP data relates to the size of memory needed to store the routing tables oin routers that operate in the Default-Free Zone, and its rates of growth ands stability can provide useful insights into the dynamics of network growth, in both IPv4 and IPv6. The dynamics of network convergence can also tell us how the BGP routing infrastructure is coping with scaling pressures. In this presentation we will dive ito metrics of BGP, comparing the data over the past two decades to the data relating to 2025.

This talk will go over Vocus' journey in the last few years about building an OOB network (yet another MPLS network), why we chose the network and hardware design and lessons we have learnt along the way.This should give participants an idea of the important of building a seperate management network, how it can be done and concerns to watch out for.

This presentation explores the practical challenges and solutions in accurately measuring campus network performance and Local Area Network (LAN) throughput using both manual and automated approaches. It features real-world case studies where unexplained slowdowns across campus buildings were traced to overlooked internal bandwidth, configuration, and latency issues. Three key tools are compared: iPerf3, a powerful manual benchmarking utility for detailed performance diagnostics; LibreSpeed, an automated, web-based solution for continuous throughput monitoring; and Smokeping, a latency visualization tool that helps detect micro-delays and instability within the LAN. The session demonstrates how combining these methods into a hybrid monitoring approach enhances end-to-end visibility across wired, fiber, and Wi-Fi segments, supports proactive capacity planning, and ensures efficient, data-driven network management.

In the old days of traditional separation of IP and DWDM, there was no need for IP folks to care about much other than what they used on their routers/switches (while the transport team ran the DWDM infra). These past DWDM infra approaches for maximizing the data capacity per fiber pair went for running more and more DWDM channels with grid spacings as small as possible. This meant that grid spacings shrank from 200GHz to 100GHz and then 50GHz with some applications even going for 25GHz.

Now in the age of IPoDWDM the bandwidth per channel keeps increasing, as complex modulation schemes come into favor over ON-OFF-Keying which has been the de facto standard so far. Those increased per channel bandwidths of 400Gbps, 800Gbps and now pushing into the 1.6Tbps realm demand for larger grids to accommodate the spectrum necessary to operate such "Superchannels". Especially the fact of coherent detection being "blind" to anything but its own wavelength has enabled interesting topologies that can omit filters altogether.

Of course that comes at a cost of reduced flexibility. Which means, IP folks now will need to understand the nuances of how coherent detection works, need for larger filters for 800Gbps and higher, and how 400GHz filters (key word here being COTS) would future proof your network by not only allowing you to grow to 1.6Tbps, but also allow backwards compatibility for legacy services on 50GHz, etc.

An update on a deck I presented at IDNOG9 in 2024. Looking at the DDoS sources and whether things have improved or not

Two real-world examples of using RIPE Atlas to diagnose routing/forwarding issues. And hints at other uses.

The paper introduces ZTP like automation using SoT tool(Netbox). Due to lack of out of band network accessibility of remote site we developed own version of automate config to new device. It was helped to faster a network rollout process and tried less human interaction. Also integrated monitoring and configuring device using python.

A step by step guide on how to build a cgnat solution using Mikrotik routers that scales, is highly resiliant and includes logging for law enforcment.

The selection of which authoritative DNS server to use by a user's DNS resolver of choice greatly determines the user experience when accessing the Internet.

In the past, some studies have looked at the behaviour of a set of implementations describing how they perform authoritative server selection and how they re-evaluate their choices in time. Other studies have leveraged the RIPE Atlas infrastructure to observe DNS resolution of a researcher-owned domain name from various points on the Internet.

This presentation extends and complement these studies by using APNIC's distributed ad system to recruit several million end-users throughout the global Internet to observe the performance of the DNS server selection process by recursive resolvers.

The presentation also profiles the DNS nameserver set for .ID domain names.

Recently DNSSEC was implemented for the .bd ccTLD and Prime Bank took the opportunity to adopt it as an early deployer. We are among the first organizations to sign a .com.bd second-level domain with DNSSEC successfully securing primebank.com.bd and myprime.com.bd.I have presented this recently on bdNOG20.

Technical standards now support domain names and email addresses in local languages. This brings new opportunities as well as promotes digital inclusion of internet users globally. This session shares the technical steps needed to host such domain names and email addresses, covering the entire journey of the registrant from registration to deployment.

This presentation shares a real-world case study of how I transformed University of Dhaka's security posture following a cyberattack in late 2024. I demonstrate how academic institutions can build enterprise-grade resilience using open-source technologies (CrowdSec WAF, Wazuh SIEM, OPNsense, Proxmox) without massive budgets.

Key topics include: incident response and recovery, implementing behavior-based WAF with community threat intelligence, SIEM integration for unified monitoring, network segmentation strategies, and cost-effective security architecture. The session covers both technical implementation details and organizational lessons learned, making it valuable for institutions facing similar budget constraints.

Attendees will learn practical approaches to security transformation, including specific tool configurations, integration patterns, and how to leverage open-source solutions for comprehensive protection. I emphasize that effective security requires strategic thinking and community collaboration rather than expensive commercial solutions.

This presentation explores the journey of achieving more than 1 Tbps throughput on a single x86 server by evolving a virtualized Broadband Network Gateway (vBNG). It walks through practical lessons learned while scaling from early multi-VM designs to a high-performance, single “Big VM” forwarder capable of extreme packet rates. The talk covers the role of SR-IOV and Intel DPDK, extensive hardware tuning, deep code-level optimizations, and major redesigns driven by CPU cache behavior and PCIe bottlenecks. Real-world performance results, including achieving 200 Mpps per CPU and scaling to 1.6 Tbps, highlight how systematic optimization, modern hardware, and iterative engineering can unlock massive throughput gains on commodity servers.

This presentation explains how we built a network bandwidth monitoring solution for our enterprise users using open-source tools. The main goal was to clearly understand overall traffic usage and see how much of the bandwidth is used for local versus internet traffic. We used LibreNMS to monitor device and interface performance through SNMP, and Akvorado to collect and analyse sFlow/NetFlow traffic data from our routers.By visualising everything through a single Grafana dashboard, the solution removes the need to switch between multiple monitoring systems, making it easier to use while improving accuracy and data security.

This presentation describes the use of the Internet Protocol stack to implement a network in deep space. The architecture is presented with a list of key needed adaptations: (1) temporary buffering of IP packets on forwarders with intermittent links, (2) configuration of the QUIC transport protocol with parameters related to the expected round-trip time and (3) configuration of application timeouts. The simulations described in this paper confirmed the suitability of the IP suite in deep space. Applications and network services are also discussed. Finally, benefits of using the Internet Protocol suite are described.

This paper talks about the current state of IPoDWDM, how, where and by whom it is being deployed, as well as its future.

RFC8806, also known as LocalRoot, provides a mechanism for resolvers to fetch, and cache a local copy of the IANA Root Zone.

This has several benefits, including increased reliability, increased performance, improved privacy, and decreased or mitigation of the effect of some types of DoS attacks.

This presentation is related to work in the IETF to expand the usage of LocalRoot (draft-wkumari-dnsop-localroot-bcp)

As networks evolve toward distributed and cloud-native architectures, precise timing has become a silent enabler of digital infrastructure. However, much of today's synchronization still relies heavily on GNSS - a single external dependency that can be affected by jamming, spoofing, or regional disruptions.

This session explores how network-based timing can strengthen resilience and autonomy by reducing or even eliminating satellite dependency. We'll look at practical approaches to achieving sub-microsecond accuracy over packet networks using PTP, enhanced boundary clocks, and advanced monitoring and verification tools.

Drawing from real deployments across Asia's telecom and data center ecosystems, the talk highlights lessons learned, architectural trends, and key challenges in building GNSS-independent synchronization environments.

Join this session to uncover the real facts behind "GNSS independence" - not as a theoretical goal, but as a practical pathway toward more secure, stable, and future-ready network timing.

This talk will recap fundamental optical concepts, make an attempt to help understand coherent pluggables and how it makes IPoDWDM possible.

I wanted to deliver this talk to engineers and sysadmins of the community to show that incident response is not magic as well as what impact AI has on infections. Lastly discussing how correctly architecting the environment impacts attack chain. The purpose of this talk is share my experience with IR cases from vulnerability and TTP perspective.

Network configuration was once a hand-crafted task when protocols were relatively simple and topologies infrequently changed. Now we're building massive, highly resilient and operationally complex networks that demand consistency at scale. During this talk we’ll take a journey through the evolution of configuration technologies and what we learned along the way. I'll cover how we started to solve these problems and how we should be configuring networks into the future. We’ll be discussing scripting, imperative operations, workflows, playbooks, APIs, RPCs, declarative models, and intent-driven operations.

The rapid evolution of Low Earth Orbit (LEO) constellations, laser inter-satellite link (LISL), and non-terrestrial networks (NTN) is reshaping global connectivity and challenging long-established assumptions about latency, routing, and rural coverage. In this presentation, I explore how next-generation satellite systems can be integrated with terrestrial mobile networks to deliver resilient, wide-area, and performance-efficient communication services.

Using real-world measurements collected through the RIPE Atlas platform, I compare end-to-end performance of Starlink LEO satellite connectivity with 4G/5G mobile networks across multiple European countries. Over a 7-day continuous measurement period, I analyzed long-haul RTT, hop-count behavior, intercontinental routing paths, and jitter distribution toward an APNIC-hosted target in Singapore. Despite longer physical paths, Starlink consistently demonstrated lower and more stable long-haul latency, enabled by backbone efficiency and evolving ISL-assisted routing, while mobile networks showed higher jitter and EPC-driven processing delays.

The talk also provides a deep-dive into satellite architecture — from LEO mesh topology, ground stations, and LISL types (intra-plane, inter-plane, crossing-plane) to practical NTN–TN interoperability considerations defined in 3GPP TR 38.811. Real-world examples illustrate how geography, beam handovers, congestion, atmospheric conditions, and PoP presence influence user experience.

Through this measurement-driven analysis, the session highlights where satellite networks outperform terrestrial systems, where limitations remain, and how future NTN–TN integration could enable seamless global connectivity — especially across underserved regions that represent 80% of the world’s landmass without mobile coverage. The study outlines future research directions, including TLS latency profiling, global LEO probe integration, and anomaly detection in transcontinental routing.

This presentation provides operators and engineers with practical insights into how terrestrial and satellite infrastructures can complement rather than compete, and what this means for the future of Internet reachability, resilience, and interconnection.

Is Artificial Intelligence a cheat, a powerful tool, or something more?

This presentation delves into AI's evolving role, drawing parallels to past technological shifts like calculators and Excel. We’ll explore the undeniable benefits of AI, from boosting productivity to enabling groundbreaking innovation, while critically examining its risks, including data hallucinations and the proliferation of misinformation. From AI's foundational purpose to the cutting-edge applications of generative AI in complex industries like network interconnection, join me to understand how this transformative technology acts as a tool, teaches new possibilities, and at times, plays the trickster, demanding an open yet cautious mind.

Most ISPs in Bangladesh rely on public DNS resolvers like Google or Cloudflare. This means operators have almost no insight into where their users’ CDN traffic is really going — which cache nodes serve them, and whether those are on-net or global.

This session shares a practical, low-cost approach using PowerDNS Recursor, ClickHouse, and Grafana to build local DNS visibility. We used live resolver logs to classify CDN answers (Udoy, Vendor, Global) for Facebook (FNA) and Google (GGC).

Attendees will learn how to instrument DNS within their own networks, interpret query-to-CDN mapping, and use the data for routing, cache deployment, and QoE improvement.

The Internet Engineering Task Force (IETF) works on developing the voluntary standards that underpin the global Internet. These standards often influence operational practices, network architecture, and protocol evolution across the ecosystem. This session will highlight some key developments at the IETF, including new standards proposals, newly chartered working groups, and hot topics gaining traction within the IETF community. Special attention will be given to areas of direct relevance to network operators.

Whether you’re tracking standards adoption, involved in deployments and operations, or just curious about emerging protocol trends, this talk will offer a concise, operator-focused view of what’s brewing at the IETF.

Public BGP data provided by collection platforms such as RIPE RIS and RouteViews is essential to the Internet community. It enables operators and researchers to monitor the global routing system and underpins many widely used tools, including BGP looking glasses and hijack-detection systems.

In this talk, we introduce bgproutes.io, a next-generation BGP data collection platform designed to improve routing visibility and usability. We explain how bgproutes.io differs from existing platforms, and how operators and researchers can access and explore the collected data through a public API and interactive analytical dashboards.

A key focus of the talk is how bgproutes.io leverages recent BGP technologies to significantly increase data coverage. In particular, we describe how the platform uses the BGP Monitoring Protocol (BMP) to collect routing information beyond best-path announcements, enabling visibility into routes received from multiple peers and at different stages of the BGP decision process. We share operational experience showing how a small number of BMP sessions already translates into a large number of additional vantage points.

We also present how bgproutes.io performs RPKI and ASPA validation on every collected route and exposes validation results to users. We argue that providing such validation data at scale can significantly help the community assess the deployment, effectiveness, and practical impact of these security mechanisms.

Overall, this talk introduces not only a new BGP data collection platform, but also demonstrates how emerging technologies such as BMP and ASPA can be deployed and used in practice to improve and facilitate Internet routing monitoring.

bgproutes.io is partially funded by the APNIC foundation.

Internet Exchange Points (IXPs) are crucial components of today’s Internet ecosystem, enabling efficient interconnection among Autonomous Systems (ASes) and handling a substantial share of global Internet traffic. Their operation relies on Route Servers (RSes), which simplify public peering by allowing ASes to maintain a single Border Gateway Protocol (BGP) session rather than multiple bilateral ones. However, BGP's reliance on implicit trust exposes it to vulnerabilities that can be exploited to hijack or disrupt traffic. To mitigate these risks, IXPs commonly employ filtering mechanisms based on Internet Routing Registries (IRRs) and the Resource Public Key Infrastructure (RPKI). Current practices exhibit a critical blind spot: IRR-based filtering heavily relies on AS-SET objects, failing to explicitly bind IP prefixes to their legitimate AS, allowing misconfigurations or hijacks to evade detection. In this work, we formally define and analyse this vulnerability, demonstrating how it can be exploited to perform prefix hijacking via IXPs. We quantify its prevalence across a large set of IXPs. We further propose practical countermeasures to strengthen RS filtering and validate our findings using real-world data from the RSes of two major IXPs: AMS-IX and NAMEX.

This presentation details the journey undertaken by IX.br in migrating the São Paulo network from LDP+VPLS technology to SRv6+EVPN. With the completion of the project, we present some results and points that are still pending.

Peering is no longer confined to a single building or a traditional Internet Exchange. Cloud, edge deployments, and automation are reshaping how networks discover, connect, and exchange traffic.

This panel brings together IXP operators, network providers, and infrastructure connectors to explore how these shifts are changing peering models across the Asia-Pacific region and beyond.

Drawing on real-world deployments and operational experience, the discussion will examine how traffic flows are moving outward from centralised hubs toward distributed, software-driven interconnection.

RouteViews was initiated in 1995 at the University of Oregon to allow network operators see what their routing tables looked like from an external perspective. Over the years since, this community project has set up collectors in strategic locations around the world, allowing network operators a rich view of their routing announcements, and researchers a valuable resource for tracking the growth and variations in the global Internet infrastructure.

This presentation introduces the new team behind RouteViews, and shares the new latest developments in the project, now operated on behalf of the global Internet community by the Network Startup Resource Center (NSRC). Some of the developments are the API, a new BMP processor, How to use the RouteViews system as a network operator and what we are working on for the future.

Indonesia’s Internet ecosystem has evolved from a single Internet Exchange Point in 1997 to more than fifty IXPs nationwide, connecting over one thousand ISPs and thousands of non-ISP networks. While this growth strengthens domestic connectivity, it also creates complex routing challenges. Many ISPs now connect to multiple IXPs in Jakarta, leading to routing table bloat, suboptimal paths, inconsistent community tagging, and uneven visibility across networks. This presentation highlights real operational issues faced by Indonesian network operators, including congestion, asymmetric routing, and manual policy management caused by the absence of shared standards and unified monitoring systems. The discussion emphasises that multi-IXP routing problems are not only technical but also organisational. Achieving a stable and efficient Internet ecosystem in Indonesia requires improved communication, coordination, and transparency among ISPs, IXPs, and the broader networking community.